CDA 3100 flashcards

Why CDA

Computer architecture performance and energy systems will be affected and tells us how computers operate

Iron Law

CPU time = instruction count CPI * clock period

Clock period

inverse of clock rate

Amdall’s law:

“dont gamble”

Moore’s Law:

states integrated circuit resources doubles every 18-24 months

History of Moore’s Law

resulted from 1965 prediction of growth in IC capacity by Gordon Moore

Moores Law Conditions

Must anticipate where the technology will be when the design finishes rather then when the design starts , represents designing for rapid change

Abstraction

represents the design at different levels of representation, lower level details are hidden to offer simpler models at higher levels

Common Case Fast

Simpler than rare case, you already know what the case is

Performance Via Parallelism

more performance means faster completion all done at the same time

Performance Via Pipelining

Split up tasks will compute faster, example from text book “firemen respond to fire , or a human chain carry a water to a fire as it is faster than individuals moving back and forth.

Performance via Prediction

it is faster is assume it is to work then wait until you know the results.

Hierarchy of Memories

The fastest, smallest and most expensive per bit at the top of the hierarchy and the slowest, largest, and cheapest per bit at the bottom.

Dependability Via Redundancy

Take over when failure occurs, help detect failures, make systems dependable including redundant components that take over when failure occurs

Idea of abstraction

to go from a complex application to the simple instructions, involving several layers of software that interpret or translate high-level operations into simple computer instructions

Operating system

interface between a users program and the hardware and provides a variety of services and supervisory functions

Compilers

translations of a program written in high level language 8

From high level language to hardware

each letter is a binary bit , send electrical systems for on and off binary bits are instructions , a collection of bits that the computer understands

Assembly Language

requires the programmer to write one line for every instruction that the computer will follow.

LCD

controls the transmission of light

active matrix

has a tiny transistor switch at each pixel to control current and sharper images. a red green blue mask associated on each dot on the display determines the intensity of three color components in the final image

bit map

matrix of bits

raster refresh

or frame buffer to store the bit map and accounts for the buffer on screen

datapath

preforms arithmetic operations

control

tells the datapath, memory and I/O devices what to do according to the instructions of the program

memory

is where programs are kept when they are running

DRAM

dynamic random access memory, multiple DRAMs are used to contain the instructions and data of a program

Cache memory

consists of a small fast memory that acts as a buffer for DRAM

SRAM

faster and less dense and hence more expensive than DRAM

Instruction set architecture

includes anything the programmer needs to know to make a binary machine language program, etc

abstraction

interface between hardware and low level instructions

transistor

simply an on off switch controlled by electricity

IC integrated circuit

combines dozens of transistors into a single chip

very large scale integrated circut

describes the increase from hundreds to millions of transistors

semi conductors

materials that does not conduct electricity well

yield

percentage of good dies from the total number of dies on the wafer

execution time

the total time required for the computer to completer a task

cost of an integrated circuit

cost per die = cost per water / die per wafter(yield)

dies per water = water area/ die area

yield = 1/ (1+(defects per area * die area/2))²

bandwidth

the total amount of work dome om a golem to,e

preformance

1/execution time

CPU time

time the cpu spends computing for the task and does not include time spent waiting for IO

clock cycles per instruction

average number of clock cycles each instruction takes to execute

CPU execution time

CPU clock cycle * clock cycle time

CPU execution time pt2

CPU clock cycles / clock rate

CPU time

instruction count* CPI* Clock cycle time

Instruction Mix

a measure of the dynamic frequency of instructions across one of many programs

CMOS

Complementary metal oxide semiconductor, the primary source of energy consumption is dynamic energy

Dynamic Energy

Energy (Capacitive load ) (Voltage²)

Energy of a single transition

Energy(1/2)(capacitive load)(Voltage²)

Power required per transisiton and frequency of transitions

Power ½ capacitive load (voltage²)(frequency switched)

Amdahl’s law

the execution time of the program after making the improvement is even by

Execution time affected by improvement/ amount of improvement + Execution time unaffected

MIPS

Millions of instructions per second MIPS=

Instruction count / execution time * 10^6